Highly active organocatalysts for asymmetric anti-mannich reactions

Article abstract of DOI:10.1002/chem.201101513

Lighten the load! A family of enantiopure 4-oxy-substituted 3-aminopyrrolidines arising from the enantioselective ring-opening of meso-3-pyrroline oxide have been developed as catalysts for the asymmetric, anti-selective Mannich reaction (see scheme; PMP=

Full text of DOI:10.1002/chem.201101513

DOI: 10.1002/chem.201101513
Highly Active Organocatalysts for Asymmetric anti-Mannich Reactions
Rafael Martꢀn-Rapffln,[a] Xinyuan Fan,[a] Sonia Sayalero,[a] Mahboubeh Bahramnejad,[a]
FØlix Cuevas,[a] and Miquel A. Pericàs*[a, b]
Aminocatalysis has undergone an intense development
over the last decade, mainly due to its potential for simplify-
ing and improving many different reactions and proce-
dures.[1] Until now, proline-derived catalysts account for
most applications in this field. These species are generally
derived from natural l-proline or trans 4-hydroxyproline,
and therefore the functionalization pattern of the pyrroli-
dine ring appears somehow restricted to a 2- or 2,4-substitu-
tion. Enantioselective Mannich reactions based on aminoca-
talysis offer the most straightforward route to diastereomeri-
cally and enantiomerically enriched functional amino
acids.[2] Whereas syn-selective Mannich reactions now repre-
sent a rather well-solved problem,[2,3] the corresponding
anti-selective processes still lack generally useful catalysts
able to process both aldehyde and ketone substrates at low
catalyst loading and under convenient reaction conditions.
The first, partially successful reports on this process were
based on (diphenyl)prolinol derivatives.[4] Only recently,
highly effective anti-selective aminocatalysts have been ra-
tionally designed by incorporating a hydrogen-bond donat-
ing group at the appropriate distance from the secondary
amine.[5,6] Barbas et al. found that a carboxylic acid group
placed at C3 on the pyrrolidine ring very efficiently controls
the enantioselectivity of the reaction.[5] An additional
methyl substituent at C5 further increases the stereoselectiv-
ity, but at the expense of nearly suppressing the reactivity
with ketones (Scheme 1a, I and II). This design has been fur-
ther developed, by replacing the carboxy directing group by
other hydrogen-donating groups (Scheme 1a, III and
IV).[5c,7]
Scheme 1. a) Selected examples of pyrrolidine-based anti-selective cata-
lysts for the Mannich reaction; b) Design of the catalysts presented in
this work.
tion of a N-protected 3-pyrroline oxide with a nitrogen nu-
cleophile[8] could provide a most simple, yet new design of a
new type of efficient organocatalysts (Scheme 1b). We wish
to report in this communication the development of a new
type of extremely active, yet highly enantioselective organo-
catalysts, according to the design outlined in Scheme 1b, for
the anti-selective direct Mannich reactions of aldehydes and
ketones.
Catalysts 1–3 were readily prepared[9] in >99.5% enantio-
meric excess (ee) through key intermediate 4, which arises
from the desymmetrization of meso-N-trifluoroacetyl-3-pyr-
roline oxide through the sequences shown in Scheme 2.
Thus, asymmetric ring-opening of N-trifluoroacetyl-3-pyrro-
line oxide with TMSN3 catalyzed by (R,R)-(salen)CrIII,[10]
followed by consecutive deprotection and N-tert-butoxycar-
bonyl (Boc) protection[11] leads to 4 with 96% ee (84%
yield). Quite interestingly, this material can be enantioen-
riched to >99.5% ee through a single recrystallization. Hy-
drogenolysis of the azide in 4 and subsequent treatment
with triflic anhydride (Tf2O) leads to the common inter-
mediate 6, in which diversity can be easily introduced. Sily-
lation and subsequent pyrrolidine deprotection gives cata-
lysts 2 and 3 (Scheme 2, reactions a–e). Catalyst 1, in turn, is
best prepared through the sequence shown in Scheme 2 (re-
actions f–h) in which the methoxy group is installed on in-
termediate 4 before manipulation of the azido group.
In light of these precedents, 4-substituted 3-aminopyrroli-
dines appeared as promising catalyst candidates for direct
aminocatalytic enantioselective anti-type Mannich reactions
of aldehydes and ketones. We reasoned that desymmetriza-
[a] Dr. R. Martꢀn-Rapffln, X. Fan, Dr. S. Sayalero, M. Bahramnejad,
Dr. F. Cuevas, Prof. M. A. Pericàs
Institute of Chemical Research of Catalonia (ICIQ)
Av. Països Catalans 16, 43007 Tarragona (Spain)
Fax : (+34)977-920-222
E-mail: mapericas@iciq.es
With these catalysts in hand, the reaction of heptanal with
preformed N-(p-methoxyphenyl) (N-PMP) ethyl glyoxylate
imine 11 was used for the optimization of reaction condi-
tions (see the Supporting Information). Very gratifyingly,
the reaction afforded with all three catalysts the desired
[b] Prof. M. A. Pericàs
Departament de Quꢀmica Orgànica, Universitat de Barcelona
08028 Barcelona (Spain)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201101513.
8780
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 8780 – 8783

COMMUNICATION
Table 1. anti-Selective Mannich reactions of aldehydes catalyzed by 1–
3.[a]
Entry
R
Cat.
[mol%]
t
Yield[b]
[%]
d.r.[c]
anti/syn
ee[d]
[%]
G
[h]
1
2
3
4
5
6
7
8
n-C5H11
n-C5H11
n-C5H11
n-C5H11
iPr
iPr
iPr
iPr
Bn
Bn
Bn
Bn
Bn
Bn
Me
Me
Me
1
2
2
3
1
2
2
3
1
2
2
2
2
3
2
2
3
2
2
0.5
0.5
0.1
0.5
0.5
0.5
0.1
0.5
0.5
0.5
0.1
0.05
0.01
0.5
0.5
0.1
0.5
0.5
0.1
1.5
2
7
2
7
7
14
7
2
3
6
8
74
2
2
7
2
1
5
84
90
88
83
84
87
88
91
85
90
89
83
67[c]
82
79
68
77
80
81
92:8
96:4
96:4
94:6
88:12
94:6
94:6
88:12
92:8
94:6
95:5
95:5
94:6
90:10
92:8
94:6
89:11
95:5
87:13
99
99
99
98
95
98
96
95
94
97
96
96
95
97
97
97
97
97
97
9
Scheme 2. Synthesis of catalysts 1–3. a) i) TMSN3 (R,R)-(salen)CrIII
,
10
11
12
13
14
15
16
17
18
19
Et2O, ii) K2CO3, iii) Boc2O, Na2CO3, 84% (96% ee) (3 steps) or 67% re-
cryst. (99.5% ee); b) H2, PtO2, MeOH, 93%; c) Tf2O, DIPEA, CH2Cl2,
À788C, 52%; d) TBS-Cl or TBDPS-Cl, imidazole, DMF, 83–97%; e) tri-
fluoroacetic acid (TFA), CH2Cl2, 80–85%. f) NaH, tetrabutylammonium
iodide (TBAI), MeI, THF, 08C, 90%; g) i) triphenylphosphine (TPP),
H2O, THF; ii) Tf2O, N,N-diisopropylethylamine (DIPEA), CH2Cl2,
À788C, 50% (2 steps); h) i) TFA, CH2Cl2, ii) Dowex 50WX8, 90% (2
steps).
9-nonenyl
9-nonenyl
[a] Conditions: 11 (0.250 mmol), catalyst (0.025–1.25 mmol), aldehyde
(0.375 mmol), EtOAc (2 mL), 08C. [b] Yield of the isolated diastereo-
meric mixture of anti- and syn-products. [c] Determined by 1H NMR
analysis of the crude. [d] Determined by chiral HPLC analysis.
anti-adduct in high yields and excellent selectivity in a varie-
ty of solvents in very short times with the use of uncommon-
ly low (<1 mol%) catalyst loadings.[12] The best reaction
conditions involved working in ethyl acetate at 08C, al-
though work at room temperature had an almost negligible
effect on both diastereoselectivity and enantioselectivity. In
view of the non-discriminating results obtained in the reac-
tion of heptanal, we decided to assess the scope of the reac-
tion using all three catalysts under the optimized reaction
conditions (Table 1). Compounds 1–3 successfully catalyzed
the reaction for a variety of aldehydes leading to anti-type
Mannich adducts in typically 1–3 h with very high diastereo-
selectivity and enantioselectivity at 0.5 mol% catalyst load-
ing. As a general trend, whereas 2 gives the best enantiose-
lectivities, catalyst 1 leads to shorter reaction times. For re-
actions catalyzed by 2, the catalyst loading was further re-
duced to 0.1 mol% (Table 1, entries 3, 7, 11, 16, and 19). We
were pleased to find that under these reaction conditions
complete conversions were recorded in most cases in less
than 7 h at 08C with practically no decrease in enantioselec-
tivity. For 3-phenylpropanal, the reaction was also conduct-
ed at 0.05 mol% and with 0.01 mol% catalyst loadings
(Table 1, entries 12 and 13).[13] Even under these unprece-
dented conditions, the enantioselectivity remained above
95% ee. Remarkably enough, these results clearly demon-
strate the superior activity of catalysts 1–3 with respect to
previously developed catalysts for anti-selective Mannich re-
actions with aldehydes. This allows for a 50–200 fold de-
crease in catalyst loading when the present standard condi-
tions (0.1 mol%) are compared with those of the current
protocols.[4,5a,c,6,7]
Catalysts 1–3 were next tested with the comparatively less
reactive ketone substrates (Table 2). In a initial series of ex-
periments, a representative set of ketones, including both
cyclic and acyclic substrates, were reacted with imine 11 in
the presence of 1–3 at 5 mol% loading at room temperature.
Under these conditions, complete conversion of cyclic sub-
strates was observed in 2–6 h, in which time the correspond-
ing adducts were obtained with very high diastereo- and
enantioselectivity. Lowering the catalyst loading to 1 mol%
(Table 2, entries 4 and 5, 12 and 13, and 16 and 17) led to in-
creased reaction times, but with only marginal erosion in the
stereochemical outcome of the reactions. Acyclic ketones re-
quired longer reaction times for complete conversion, but
very high enantioselectivities were still recorded (Table 2,
entries 18 and 19). As already noted for the reaction with al-
dehydes, the use of catalysts 1–3 involves very convenient
conditions (i.e., reactions at room temperature with practi-
cally no excess of ketone reactant)[14] and allows a reduction
of nearly one order of magnitude in the amount of catalyst
used for the anti-selective Mannich reaction of ketones in
standard protocols,[5b,7a,b,d] while the recorded enantioselec-
tivities are among the best for this reaction.
The role exerted by a C4 substituent at the pyrrolidine
ring in increasing the catalytic activity and the diastereose-
lectivity, such as in systems 1–3 with respect to a C5 substi-
tuted system such as I (Scheme 1a), can be tentatively ra-
tionalized through the schematic representation shown in
Scheme 3. A C5 substituent, while beneficial for the confor-
Chem. Eur. J. 2011, 17, 8780 – 8783
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8781

M. A. Pericàs et al.
Table 2. anti-Selective Mannich reactions of ketones catalyzed by 1–3.[a]
Entry Substrate
Cat.
[mol%]
t
Yield[b] d.r.[c]
ee[d] [%]
G
[h]
[%]
anti/syn
1
1
5.0
6
6
92
92
95:5
96
2
3
4
5
2
3
2
3
5.0
5.0
1.0
1.0
96:4
95:5
96:4
96:4
98
98
94
94
4.5 89
20
24
90
89
Scheme 3. Effect of a) a C5 substituent and b) a C4 substituent on the
conformational behavior of enamines derived from pyrrolidine-type orga-
nocatalysts.[14]
6
1
5.0
2
81
96:4
92
7
8
2
3
5.0
5.0
3.5 78
95:5
83:17
93
83
6
72
pyrrolidines 1–3 that very efficiently catalyze the direct,
anti-selective Mannich reaction of aldehydes and ketones.
Unprecedented levels of activity have been determined for
the new catalyst type, which allows using catalyst loadings as
low as 0.01 mol% under very convenient reaction condi-
tions. This new family of organocatalysts, whose activity
characteristics step into a land normally reserved to metal
catalysis, is now being integrated into reaction systems for
process intensification in our laboratories.
9
1
5.0
2
86
99:1
95
10
11
12
13
2
3
2
3
5.0
5.0
1.0
1.0
3.5 84
4.5 82
20
24
97:3
98:2
97:3
96:4
93
94
88
93
82
83
14[e]
1
5.0
6
62
80:20
96
15[e]
16[e]
17[e]
2
2
3
5.0
1.0
1.0
5.5 60
20
24
78:22
81:19
75:25
94
87
89
56
52
Experimental Section
18[e]
2
2
5.0
5.0
28
20
70
86:14
76:24
96
96
19[e]
68[f]
General procedure (Table 1): The reactions in Table 1 were performed in
a closed vial under atmospheric conditions. Protected a-imino ester 11
(0.25 mmol, 1.0 equiv) and the appropriate catalyst (1–3) were dissolved
in ethyl acetate (1.0 mL). Aldehyde (0.37 mmol, 1.5 equiv) was then
added to the solution. After stirring at 08C for the indicated time in
Table 1, the reaction mixture was concentrated under reduced pressure
and purified by flash column chromatography. In general the anti- and
syn isomers of the Mannich products shown in Table 1 were not separat-
ed by column chromatography.
[a] Conditions: 11 (0.250 mmol), catalyst (0.0025–0.0125 mmol), ketone
(0.375 mmol), EtOAc (2 mL), RT. [b] Yield of the isolated diastereomeric
mixture of anti- and syn-products. [c] Determined by 1H NMR or HPLC
analysis of the crude. [d] Determined by chiral HPLC analysis. [e] 10
equivalents of ketone were used. [f] Containing approximately 10% of
regioisomer.
General procedure (Table 2): The reactions in Table 2 were performed in
a closed vial under atmospheric conditions. Compound 11 (0.25 mmol,
1.0 equiv) and the appropriate catalyst (1–3) were dissolved in ethyl ace-
tate (1.0 mL). Ketone (0.37 mmol, 1.5 equiv for entries 1–13; and
2.50 mmol, 10 equiv for entries 14–19) was then added to the solution.
After stirring at RT for the indicated time in Table 2, the reaction mix-
ture was concentrated under reduced pressure and purified by flash
column chromatography. In general the anti- and syn isomers of the Man-
nich product shown in Table 2 were not separated by column chromatog-
raphy.
mational control of the intermediate enamines with alde-
hyde substrates (R2 =H, Scheme 3a),[15] leads to a decrease
in the concentration of the reactive enamine conformer with
ketone substrates (R2 =alkyl, Scheme 3a) and likely pro-
vokes unfavorable steric interactions between reactants in
the transition state leading to the anti adduct. As a result,
catalyst I is inefficient with ketone substrates. On the con-
trary, an R-oxy substituent at C4 on the pyrrolidine ring
(Scheme 3b) seems to efficiently control the conformation
of the enamine intermediate, favoring the reactive confor-
mer even with ketone substrates. According to this interpre-
tation, and to the experimental results presented here, a
trans-3,4-disubstitution pattern in pyrrolidines, with one
group acting as hydrogen-bond activator and the other as
steric controller of enamine conformation, appears to be op-
timal for catalysis in the anti-selective Mannich reaction.
In summary, we have used a desymmetrization strategy to
convert an inexpensive N-protected 3-pyrroline oxide into a
small family of enantiopure 4-alkoxy and 4-silyloxy 3-amino-
Acknowledgements
This work was funded by MICINN (Grant CTQ2008-00947/BQU, Con-
solider Ingenio 2010 Grant CSD2006-0003 and Juan de la Cierva Fellow-
ship (R.M.R.)), DIUE (Grant 2009SGR623), and ICIQ Foundation.
Keywords: asymmetric catalysis
·
desymmetrization
·
mannich reactions · organocatalysis · ring-opening
8782
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 8780 – 8783

Highly Active Organocatalysts for Asymmetric anti-Mannich Reactions
COMMUNICATION
[6] a) T. Kano, Y. Yamaguchi, O. Tokuda, K. Maruoka, J. Am. Chem.
Soc. 2005, 127, 16408–16409; b) T. Kano, Y. Yamaguchi, K. Maruo-
ka, Angew. Chem. 2009, 121, 1870–1872; Angew. Chem. Int. Ed.
2009, 48, 1838–1840; c) T. Kano, Y. Yamaguchi, K. Maruoka, Chem.
Eur. J. 2009, 15, 6678–6687.
[7] a) M. Pouliquen, J. Blanchet, M.-C. Lasne, J. Rouden, Org. Lett.
2008, 10, 1029–1032; b) T. Kano, Y. Hato, K. Maruoka, Tetrahedron
Lett. 2006, 47, 8467–8469; c) T. Kano, Y. Hato, A. Yamamoto, K.
Maruoka, Tetrahedron 2008, 64, 1197–1203; d) H. Zhang, Y. Chuan,
Z. Li, Y. Peng, Adv. Synth. Catal. 2009, 351, 2288–2294; e) E.
Gómez-Bengoa, M. Maestro, A. Mielgo, I. Otazo, C. Palomo, I. Ve-
lilla, Chem. Eur. J. 2010, 16, 5333–5342.
[8] E. N. Jacobsen, M. H. Wu in Comprehensive Asymmetric Catalysis
(Eds: E. N. Jacobsen, A. Pfaltz, H. Yamamoto), Springer, New
York, 1999, Chapter 35.
[9] Complete details for the preparation of 1–3 are given in the Sup-
porting Information.
[10] a) L. E. Martꢀnez, J. L. Leighton, D. H. Carsten, E. N. Jacobsen, J.
Am. Chem. Soc. 1995, 117, 5897; b) S. E. Schaus, J. F. Larrow, E. N.
Jacobsen, J. Org. Chem. 1997, 62, 4197.
[11] A. I. Oliva, U. Christmann, D. Font, F. Cuevas, P. Ballester, H.
Buschmann, A. Torrens, S. Yenes, M. A. Pericàs, Org. Lett. 2008, 10,
1617–1619.
[12] The reaction was also explored with other types of imines. When
the N-Boc imine of benzaldehyde was reacted with propanal in the
presence of 2 (5 mol%), either in ethyl acetate or in chloroform fol-
lowing the procedure described in ref. [4e], complete conversion
was recorded in 44 h at RT. The best selectivity was achieved in
chloroform, with anti/syn=57:43 (ee (anti): 72%; ee (syn): 75%). In
view of the results, which are in line with previous observations by
Barbas et al (ref. [5c]), we decided to focus on N-PMP ethyl glyoxy-
late imine 11.
[13] Maruoka et al. (ref. [6a]) reported a single case where the reaction
of propanal with the PMP imine of ethyl glyoxylate was catalyzed
by 0.2 mol% of an axially chiral aminosulfonamide; the reaction re-
quired 22 h at RT for completion.
[1] a) A. Berkessel, H. Grçger, Asymmetric Organocatalysis: From Bio-
mimetic Concepts to Applications in Asymmetric Synthesis, Wiley-
VCH, Weinheim, 2005; b) Enantioselective Organocatalysis: Reac-
tions and Experimental Procedures (Ed.: P. I. Dalko), Wiley-VCH,
Weinheim, 2007; c) P. Melchiorre, M. Marigo, A. Carlone, G. Barto-
li, Angew. Chem. 2008, 120, 6232–6265; Angew. Chem. Int. Ed.
2008, 47, 6138–6171; d) X. Yu, W. Wang, Org. Biomol. Chem. 2008,
6, 2037–2046; e) M. Marigo, P. Melchiorre, ChemCatChem 2010, 2,
621–623.
[2] For a pioneering three-component organocatalytic Mannich reac-
tion, see: a) W. Notz, K. Sakthivel, T. Bui, G. F. Zhong, C. F. Barba-
s III, Tetrahedron Lett. 2001, 42, 199–201; b) B. List, J. Am. Chem.
Soc. 2000, 122, 9336–9337; c) B. List, P. Pojarliev, W. T. Biller, H. J.
Martin, J. Am. Chem. Soc. 2002, 124, 827–833; For recent reviews
on organocatalytic Mannich reactions, see: d) J. M. M. Verkade,
J. C. v. H. Lieke, P. J. L. M. Quadflieg, F. P. J. T. Rutjes, Chem. Soc.
Rev. 2008, 37, 29–41; e) A. Ting, S. E. Schaus, Eur. J. Org. Chem.
2007, 5797–5875.
[3] For selected examples on the aminocatalytic asymmetric syn-Man-
nich reactions, see: a) A. Córdova, S. Watanabe, F. Tanaka, W. Notz,
C. F. Barbas III, J. Am. Chem. Soc. 2002, 124, 1866–1867; b) Y. Hay-
ashi, W. Tsuboi, I. Ashimine, T. Urushima, M. Shoji, K. Sakai,
Angew. Chem. 2003, 115, 3805–3808; Angew. Chem. Int. Ed. 2003,
42, 3677–3680; c) W. Notz, F. Tanaka, S. Watanabe, N. S. Chowdari,
J. M. Turner, R. Thayumanavan, C. F. Barbas III, J. Org. Chem.
2003, 68, 9624–9634; d) E. Alza, C. Rodrꢀguez-Escrich, S. Sayalero,
A. Bastero, M. A. Pericàs, Chem. Eur. J. 2009, 15, 10167–10172.
[4] a) A. Córdova, C. F. Barbas III, Tetrahedron Lett. 2002, 43, 7749–
7752; b) J. FranzØn, M. Marigo, D. Fielenbach, T. C. Wabnitz, A.
Kjærsgaard, K. A. Jørgensen, J. Am. Chem. Soc. 2005, 127, 18296–
18304; c) I. Ibrahem, A. Córdova, Chem. Commun. 2006, 1760–
1762; d) C. Gianelli, L. Sambri, A. Carlone, G. Bartoli, P. Mel-
chiorre, Angew. Chem. 2008, 120, 8828–8830; Angew. Chem. Int.
Ed. 2008, 47, 8700–8702; e) P. Galzerano, D. Agostino, G. Benciven-
ni, L. Sambri, G. Bartoli, P. Melchiorre, Chem. Eur. J. 2010, 16,
6069–6076.
[14] Previously reported procedures use larger reagent excesses (2:1 to
10:1) for reactions of comparable substrates (see refs. [5b], [7a],
[7b], and [7d]) and/or involve inconvenient reaction conditions (see
ref. [7a]).
[5] a) S. Mitsumori, H. Zhang, P. H.-Y. Cheong, K. N. Houk, F. Tanaka,
C. F. Barbas III, J. Am. Chem. Soc. 2006, 128, 1040–1041; b) H.
Zhang, M. Mifsud, F. Tanaka, C. F. Barbas III, J. Am. Chem. Soc.
2006, 128, 9630–9631; c) H. Zhang, S. Mitsumori, N. Utsumi, M.
Imai, N. Garcia-Delgado, M. Mifsud, K. Albertshofer, P. H.-Y.
Cheong, K. N. Houk, F. Tanaka, C. F. Barbas III, J. Am. Chem. Soc.
2008, 130, 875–886.
[15] Scheme 3a has been adapted from refs. [5a–c].
Received: May 16, 2011
Published online: July 6, 2011
Chem. Eur. J. 2011, 17, 8780 – 8783
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8783